TIARA: Multi-grained Retrieval for Robust Question Answering over Large Knowledge Base

Shu, Yiheng; Yu, Zhishui; Li, Yuhan; Karlsson, Börje F.; Ma, T.; Qu, Yuzhong; Lin, Chin-Yew

doi:10.18653/v1/2022.emnlp-main.555

Cited by 15 publications

(7 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The pipeline of such systems can be broken down into two main steps: in the first step, a retriever processes an input question to gather information relevant to the formation of the corresponding logical form; in the second step, a reader, which is often a fine-tuned LM, takes in the given question and the retrieved information and outputs the desired logical form. Different systems design for different kinds of information to be retrieved and fed into the reader, such as entities and relations detected from input questions, ,, candidate logical forms, candidate query paths, , or linearized facts . To ensure that the generated queries are compliant with the ontology of the KG, many of these systems impose decoding constraints on the reader or perform an additional step of revision to realign output logical forms to the KG’s ontology. , …”

Section: Related Workmentioning

confidence: 99%

“…Different systems design for different kinds of information to be retrieved and fed into the reader, such as entities and relations detected from input questions, ,, candidate logical forms, candidate query paths, , or linearized facts . To ensure that the generated queries are compliant with the ontology of the KG, many of these systems impose decoding constraints on the reader or perform an additional step of revision to realign output logical forms to the KG’s ontology. , …”

Section: Related Workmentioning

confidence: 99%

“…More is left to be explored with semantic parsing-based approaches and the use of PLMs. Key approaches to KGQA that have benefited from generative PLMs include end-to-end translation, − the retriever-reader model, − and dynamic logical form induction. , Even though many of these systems have clinched state-of-the-art results on public benchmarks, ,, attempts to quantify their speed to demonstrate practical usage are still inadequate; we are only able to find reports of system speed measured on GPU in two papers. , This casts doubt on whether systems with impressive accuracy can be realistically deployed in a real-world setting, where specialized hardware might not be available. More broadly, we notice that efforts to study the accuracy-latency trade-off for KGQA in a hardware-constrained setting are lacking despite the wealth of similar work in other machine learning domains. − …”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Leveraging Text-to-Text Pretrained Language Models for Question Answering in Chemistry

Tran,

Pascazio,

Akroyd

et al. 2024

ACS Omega

View full text Add to dashboard Cite

In this study, we present a question answering (QA) system for chemistry, named Marie, with the use of a text-to-text pretrained language model to attain accurate data retrieval. The underlying data store is "The World Avatar" (TWA), a general world model consisting of a knowledge graph that evolves over time. TWA includes information about chemical species such as their chemical and physical properties, applications, and chemical classifications. Building upon our previous work on KGQA for chemistry, this advanced version of Marie leverages a fine-tuned Flan-T5 model to seamlessly translate natural language questions into SPARQL queries with no separate components for entity and relation linking. The developed QA system demonstrates competence in providing accurate results for complex queries that involve many relation hops as well as showcasing the ability to balance correctness and speed for real-world usage. This new approach offers significant advantages over the prior implementation that relied on knowledge graph embedding. Specifically, the updated system boasts high accuracy and great flexibility in accommodating changes and evolution of the data stored in the knowledge graph without necessitating retraining. Our evaluation results underscore the efficacy of the improved system, highlighting its superior accuracy and the ability in answering complex questions compared to its predecessor.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Leveraging Text-to-Text Pretrained Language Models for Question Answering in Chemistry

Tran,

Pascazio,

Akroyd

et al. 2024

ACS Omega

View full text Add to dashboard Cite

show abstract

“…+ denotes equal contribution 2021; Yu et al, 2022;Gu and Su, 2022;Shu et al, 2022) have achieved remarkable performance on the Zero-shot split giving the impression that KBQA generalization might be a solved problem. However, a cross-dataset evaluation of the models trained on GrailQA reveals that they do not transfer well even for the more simpler one or twohop questions.…”

Section: Introductionmentioning

confidence: 99%

The Fourth Workshop on Insights from Negative Results in NLP

2023

View full text Add to dashboard Cite

Amherst. His research focuses broadly on designing machine learning models for discourse-level language generation (e.g., for story generation and machine translation), and his group also works on tasks involving creative language understanding (e.g., modeling fictional narratives and characters). He is the recipient of best paper awards at NAACL (2016, 2018) and a best demo award at NeurIPS 2015, and he

show abstract

“…For the entity linking, we adopt the same setting as our baseline models. On GrailQA, we use the entity linking results from TIARA (Shu et al 2022). On WebQSP, we adopted the entity linking results from ELQ (Li et al 2020).…”

Section: E Implementation Detailsmentioning

confidence: 99%

Building Chinese interdisciplinary research centers : the case of Tsinghua University

Li¹,

李蓉暉²

View full text Add to dashboard Cite

Graph machine learning has been extensively studied in both academia and industry. However, in the literature, most existing graph machine learning models are designed to conduct training with data samples in a random order, which may suffer from suboptimal performance due to ignoring the importance of different graph data samples and their training orders for the model optimization status. To tackle this critical problem, curriculum graph machine learning (Graph CL), which integrates the strength of graph machine learning and curriculum learning, arises and attracts an increasing amount of attention from the research community. Therefore, in this paper, we comprehensively overview approaches on Graph CL and present a detailed survey of recent advances in this direction. Specifically, we first discuss the key challenges of Graph CL and provide its formal problem definition. Then, we categorize and summarize existing methods into three classes based on three kinds of graph machine learning tasks, i.e., node-level, linklevel, and graph-level tasks. Finally, we share our thoughts on future research directions. To the best of our knowledge, this paper is the first survey for curriculum graph machine learning.

show abstract

TIARA: Multi-grained Retrieval for Robust Question Answering over Large Knowledge Base

Cited by 15 publications

References 0 publications

Leveraging Text-to-Text Pretrained Language Models for Question Answering in Chemistry

Leveraging Text-to-Text Pretrained Language Models for Question Answering in Chemistry

The Fourth Workshop on Insights from Negative Results in NLP

Building Chinese interdisciplinary research centers : the case of Tsinghua University

Contact Info

Product

Resources

About